Abstract:
THE IMPROVElMENT OF REPRODUCTIVE PERFORMANCE B. J. RESTALL* I. INTRODUCTION Little argument need be advanced to justify effort being expended to improve the reproductive performance of our domestic animals. The history of animal husbandry has been one of continued effort to control and manipulate the animal, presumably as a consequence of man's inherent urge to interfere with all and sundry. Watson (1966) pointed out that the rate of progress in the improvement of reproductive performance in the sheep industry was small and that the reasons for it were obscure. Further, it is obvious that current reproductive performances are well below potential. In our present agricultural economy, this situation is one of concern for the reason that the rate of natural increase in stock numbers sets a limit to the rate at which scientific advances can be translated into increased production (Bishop 1964). This review will not attempt a critical examination of accrued and current research work concerned with reproduction. Space will not permit such a treatment and the question of improving reproductive performance is more complex than just considering the research problems. Our overall approach to the problem of improving reproductive rates needs to be examined in order that we might utilise our limited resources most effectively. This will be attempted in the following discussion and most of the remarks and examples used will refer to sheep, but it is hoped that those involved with cattle and pigs may gain something from such discussions. II. A GENERAL APPROACH TO THE PROBLEM Broadly, there are two ways in which reproductive performance may be improved:(a) We may aim at defining and solving existing industry problems and so bring our flocks to their existing potential. (b) We may attempt to push at the limits of performance, by investigating new ways of controlling and manipulating our animals. Both involve laboratory and field research and will involve ultimate translation into farming practice. But because we have limited resources, the very important question of priorities arises, as a poor selection of priorities can be a major * Department of Agriculture, Sydney, New South Wales. 297 limiting factor to rapid improvement. In the past, the choice of an area for research has been left largely to the individual research worker, who is probably poorly equipped and positioned to make choices that will have widespread impact. However, in the last decade, the Standing Committee for Agriculture has tackled the problem of priorities by convening two Australian Sheep and Wool Conferences (in 1963 and in 1968) and one Australian Beef Cattle Review Conference (in 1969) all of which were concerned with producing a list of priorities for the guidance of research institutions and individuals. The effectiveness of these bodies may be judged from the two Sheep and Wool Conferences. At both conferences, the determined priorities were much the same. Reproductive wastage, including failure to mate, returns to service, failure to lamb and mortality among lambs, received top billing. Further down the list were nutrition and reproduction, mating behaviour and sub-clover infertility. The lists were long ones and there was something for everybody. During the period since the first conference, a vast amount of information was obtained on all these priorities both in the laboratory and the field, but reproductive performance did not improve markedly. This situation exists despite reports at the second conference that farmers were adopting practices such as using the optimum time of mating, improving the nutrition of breeding ewes before mating and during pregnancy, and increasing lambing supervision to prevent lamb losses. It is difficult to fault the priority lists given, but something is still wrong. We may blame poor extension services, but a statement bv Watson (1966) strikes nearer the truth. He concluded that full attainment of potential performance would be hindered until field services to help individual producers were expanded, and until much more information was available on the factors which determine reproductive function in the widely varied field situations that exist, and the mechanisms whereby they do so. Perhaps our first and most important priority is to determine ways and means of doing this in a manner that will have widespread application. This leads us to consider the manner in which we conduct our research. III. RESEARCH ORGANISATION The sort of research we are considering falls into three broad categories:(a) Basic Research. The collection of information on the fundamental mechanisms controlling reproduction, and on the manner in which limiting factors may operate. All of this research can be carried out without involving commercial producers, but the information so gained is of little use unless it is taken further. (b) Relevance Research. Relevance has two components, truth and usefulness. Any information obtained from basic research must be shown to be true for any particular situation or modified accordingly. As well, it must offer some advantage to the user. Even if the question of relevance is satisfied, the commercial producer may still not be able to implement the new knowledge. (c) Feasibility Research. A piece of information or a technique may be very relevant but be of no use because it is impractical or uneconomic. If its potential value is high, research may be undertaken to find ways to make it practical. The second and third categories of research intimately and necessarily involve the commercial producer. For the ultimate application of new knowledge, not 298 only do we require research workers of high skill and training both in the laboratory and the field, but we need the effort and involvement of the individual producers. One reason for the lack of implementation, of existing information is that it is rarely in a form that permits widespread application. Local evaluation and feasibility research is clearly under-emphasised. -Our attack on the problem of raising reproductive performance on a large scale requires reorganisation at the field level and I suggest this is the first and most important priority. A lot of necessary, associated research on many facets of reproduction will be required and an attack on this first priority should quickly highlight other areas for investigation. This conclusion is not intended to underrate basic research which, as indicated above, is- an integral part of the research progression. However, basic research will be largely wasted if the information gained is not effectively implemented. Efforts are being made to come to grips with the field problem. Two examples come to mind and no doubt there are others of which I am unaware, but these two examples will illustrate the approach envisaged. Both are concerned with the Australian Sheep and Wool Conferences' top priority, reproductive wastage, and are thus concerned with improving the current situation, the first mentioned of the two broad ways in which performance may be improved. IV. INVESTIGATING THE PROBLEMS OF THE PRODUCER In 1966, officers of the N.S.W. Department of Agriculture recognised the need to attack reproductive problems in an organised manner where they occur on the commercial property. Under the guidance of Dr. Fowler, a Fertility Service was organised at the Agricultural Research Station at Trangie. This service aims at identifying sources of reproductive wastage on commercial properties and, where possible, at providing solutions to those problems either directly or by instigating m research that will provide solutions. Now in its third year of operation, a team of four investigators led by Mr. R. Connors is currently investigating 25 properties. The methods of this service are constantly under review and, while sufficient techniques are not vet available to readily identify a11 sources of loss, important areas of wastage are being determined. As well, the service provides an excellent extension vehicle for introducing new information and methods to the producer. While this service is still developing and will continue to develop for some time, it has already demonstrated its usefulness and plans are underway to set up similar services on a regional basis. Such a service demands a skilled and enterprising research backing (it has already demonstrated its capacity to outstrip research efforts) and one of its benefits will be to orientate the Department's efforts into productive avenues. It has already emphasised the need for research effort at all of the previously mentioned levels and this has led to co-operative projects with both the C.S.I.R.O. and commercial producers. The secondary priorities are showing up. A similar development has been underway in Victoria for a longer period. The excellent work of Dr. Watson and his colleagues on reproductive wastage has led to an intensification of regional research especially in the area of post-natal lamb losses. The Victorian Department of Agriculture has located research and 299 extension officers in teams at various region al centres and these teams carry out problem identification and problem solving research in close co-operation with commercial producers. Again, these teams are drawing on Research Station staff, suggesting priorities and utilising the specialist skills available there. They too have forged links with basic research through the University of Melbourne. Both of these states are still developing their approaches and methodology, but it is already clear that real Drogress can be made. it is to be hoped that these examples, together with a recognition of the progression of research information and the necessary involvement of the commercial producer, will lead to the increase in emphasis and financial backing that this type of research deserves. It might be profitable at this stage to explore some of the probIems that have arisen in these field operations. The Trangie Fertility Service has found it impossible to evaluate ram fertility from an examination of a semen sample. However, ram infertility may be indicated if returns to service of young ewes are higher than in mature ewes. Maiden ewes are apparently more sensitive to male infertility than older sheep (Connors 1970). It is commonly observed that about 25 per cent of ewes will return to mate after their first service. In a three cycle joining period, however, most ewes are settled. It thus seems unlikely that these high initial return rates indicate infertility in the ewe. More likely, they are the consequences of the mating strategy employed. The Fertility Service at the present time lacks a ready technique for estimating early embryonic loss, but it has been found that an examination of fortnightly mating records for individual ewes will indicate an extended return interval. Research is underway to determine whether this technique is a useful measure of early embryonic loss. Accurate estimates of lamb losses depend on identifving twin bearing ewes and those ewes which have lambed. The techniques described by Dun (1963) for identifying ewes that had lambed have been found to be time consuming and subject to error. The vulva1 closure technique described by Murray (1969) appears to be extremely accurate in identifying ewes that have lambed, but it is likely to have a low level of accentance on commercial properties. As yet, no technique is available for accurately determining twin bearing ewes although investigations involving the use of X-ray equipment and the measurement of progesterone levels in peripheral blood are promising. Despite these deficiencies in technique, lamb losses have been found to be a major source of wastage. It is often difficult to deduce the cause of the loss other than to implicate starvation and mismothering, probably the result of management practices. Because these problems are impairing progress in reducing reproductive wastage at the commercial level, it is beneficial to consider them from a research point of view and explore some of the methodology that may be usefully employed. (a) Evaluation of Fertility in the Male Despite the and biochemical attempts at field brack 1962), it is ejaculate. There enormous amount of knowledge concerning the morphological parameters of the male ejaculate (Mann 1964) and the many evaluation of semen samples (for example, Hulet and Ercanstill not possible to predict fertility from an examination of the are probably many reasons why this is so. Firstly, where such studies have been carried out using flock matings, the interactions of ram and ewe, season and age, to name a few variables, have rarely all been taken into account. Where positive correlations between semen narameters and fertility have been obtained in these situations, they are of little predictive value as the complex of interacting factors may be quite different in other flocks at other times. The sensitivity of the test situation to changes in a particular semen parameter will vary from flock to flock and from time to time. Secondly, the distributions of semen parameters have rarely been described. Where this has been done (Van Duijn 1965), these distributions have not been found to be normal. Thus, the use of conventional statistical methods without a proper transformation could lead to biased results. Artificial insemination (A.I.) offers a better test situation and has been used widely. However, even in this area, work has been concentrated on the ejaculates with generally 'good' semen characteristics. Even the A.I. test situation may be too insensitive to detect differences between samples with, say, motility scores of 7 and 10. In general, only these upper ranges have been examined where the resulting fertility differences might be expected to be small. As well, results derived from specific test situations would not be likely to have general application. Even if a truth were uncovered, other test situations may suppress its expression leading to a lack of confidence in the original truth. However, the approach adopted in one series of studies merits attention. Van Duijn (1965) derived, from first principles, a mathematical function describing the relationship between some semen characteristics and fertility. This function was based on two assumptions which would seem to have general applicability. The mathematical expression so derived indicated that a measure of sperm number and of their longevity, when considered in the relationship suggested, would lead to a prediction of fertility. He tested the relationship using A.I. in cattle and found very high and significant correlations. Subsequently, he examined data published by others and showed that when his formula was applied, similar high and significant correlations were present where none were thought to exist. If the assumptions behind such an approach were sound, then a generally applicable formula could be expected to result. This interesting approach is recommended for consideration by all those concerned with semen evaluation. While such a formula may operate well in an A.I. situation, the basic assumptions are likely to be inadequate for evaluating the semen of males involved in paddock matings, but they may well be extended to cover a wide range of mating practices. (b) Mating Strategy Early reports on the relationship between the numbers of ewes joined per ram and fertility indicated that one ram per hundred ewes was adequate provided small paddocks were used (Haughey 1959). No difference was found in ewe fertility following the use of one, two or three rams per hundred ewes (Edgar 1965). However, a recent report by Lightfoot and Smith (1968) has shown that fertility is responsive to manipulations of the ram to ewe ratio, at least in certain situations. These studies showed that increases in the ram to ewe ratio could lead to more ewes lambing and a high percentage of twins. Interactions were found between age of ram and the ram to ewe ratio. The ewe flocks used had grazed 301 oestrogenic clover and may well have been on a threshold of fertility that was sensitive to such changes. Nevertheless, this work indicates that manipulations of mating management may lead to beneficial results. Many'factors again are involved in this complex situation; age of ram and ewe, libido, dominance and length of mating, all of which have received attention at one time or another (Lindsay and Robinson 1961a, b; Pepelko and Clegg 1965a, b; Mattner, Braden, and Tumbull 1967; Bourke 1967; and others). Such a complex of variables is difficult to investioate; a factorial experiment designed to investigate them all would be prohibitive, Gut the situation may be aided by adopting a mathematical modelling approach in conjunction with traditional experimentation. Much of our present experimentation is analytical in the sense that it tends to break complex systems into smaller components for investigation. At best, the interactions of only two or three factors are examined at one time. Fitting together the pieces of such a jigsaw puzzle is left to the researchers' fertile mind. On the other hand, the use of mathe-matical modelling procedures is an integrative process as it puts all the variables together in an attempt to describe their true relationship. Simulation experiments then indicate the effects of changes in one or more of the variables on the whole complex. Such a modelling approach might be used with benefit in the mating situation if it is used in conjunction with animal experiments. Knowledge gained from animal experiments could be used to formulate and progessively improve the mathematical model. In turn, simulation experiments would suggest those variables likely to lead to productive results when examined in physical experiments. Manipulation o,f mating strategy offers exciting prospects and future research should be very fruitful. The use of an integrative approach as outlined above should complement the current analytical experiments and may lead to an earlier understanding of the mating complex. A word of caution - mathematical conclusions are not necessarily biological ones, and the final interpretation of results from modelling procedures can be a quicksand for the unwary biologist. (c) Early Embryonic Loss Early embryonic death in normally fertile domestic animals is substantial. Estimates of this loss vary from 15 to 21 per cent in the cow, 23 to 50 per cent in the sow (Boyd 1965) and 6 to 48 per cent in the ewe (Edey 1969). Little success in alleviating this loss has been achieved despite a vast amount of research into limiting factors presumed to operate through the mother (Edey 1969). The problem is complicated by an unknown genetic component of loss, acting independently or in conjunction with maternal effects. Bishop (1964) advanced the theory that an unsuspectedly large part of embryonic death is attributable to genetic causes, not necessarily inherited by the parents but arising de yu2vo in each parent generation and/or in the definitive -gametes. He concluded that a considerable part of embryonic death is unavoidable and should be regarded as a normal means of eliminating unfit genotypes in each generation. This raises the very important question of whether we should be looking for ways of reducing embryonic loss, a question which won't be answered until we can properly estimate the maternal contribution. 302 Biggers (1968) considered this problem and concluded, after a short mathematical exercise, that estimation of the uterine contribution to embryonic loss is impo,ssible without knowledge of the contribution of inherently abnormal embryos. He assumed in his calculations that maternal and genetic factors operate indepedently, an assumption unlikely to be true and one which has been recently explored by Giles (1970). However, the priority for research in this field is clear. Considerable reward awaits the person who finds a way to estimate the genetic component of embryonic loss. Until this is done, we cannot have much confidence in studies which assume all loss is due to- maternal factors. . (d) Lamb Mortality Neonatal losses are severe and widespread (Australian Sheep and Wool Conference Handbook 1968) and generally result from behavioural and environmental factors (Alexander, Peterson and Watson 1959). Efforts to reduce this loss have led to the development of closely supervised lambing systems (Beggs and Campion 1966; Moore, McDonald and Messenger 1966; Dun 1966; Watson et al. 1967; Giles 1968; Watson et al. 1968). In Victoria, where lamb losses are largely due to weather (Alexander, Peterson and Watson 1959; Watson and Elder 1961), lambing systems providing some form of shelter have proven beneficial, but it is more difficult to evaluate intensive lambing systems designed to avoid mismothering starvation. Despite reports that intensive lambing systems result in increases of 20 to 30 per cent in lambing percentages (Beggs and Campion 1966), it is not clear whether this gain is due to the actual lambing system or to improvement in other husbandry skills which have also changed. Several of these intensive systems have elaborate 'let out' procedures for the ewes and lambs (Watson et aZ. 1967) suggesting that intensive handling can itself contribute to losses unless care is taken. The drift system of lambing used at the Agricultural Research Station at Trangie, N.S.W. (Dun 1966) was shown to have an advantage when compared to a 'normal' station lambing. This 'control' lambing involved frequent mustering and mothering procedures. Recently, Giles at Trangie divided pregnant ewes into weekly mating groups and left them totally undisturbed until approximately one week after lambing ceased. These undisturbed ewes had the same magnitude of lamb loss (11 to 12 per cent) as the ewes in the drift system. These results indicate that we need more information before specific recommendation as regards the optimum lambing system for any region-can be made. While it does appear that refinements to an unsupervised large paddock system of lambing management can reduce lamb losses, the most economical lambing strategy for some producers may prove to be the taking of a vacation. INCREASING THE POTENTIAL FOR REPRODUCTIVE PERFORMANCE This discourse has largely ignored the vast amount of information accruing on the basic physiological mechanisms underlying reproduction, and it remains to briefly indicate how application of this knowledge in several areas of applied research may enlarge the potential of our animals for reproductive performance. 303 V. Some of the most exciting advances on the basic research of reproduction are those concerned with the measurement of pituitary hormones levels in blood (Midgley 1967; Goding et al. 1969; Wheatley and Radford 1969) and the elucidation of central nervous system control of pituitary function (Radford 1966). The techniques and information emanating from research in these fields are providing new avenues for research in the following areas:(a) The selection of animals of high reproductive potential It has been established that selection for multiple births leads to increases in reproductive performance (Turner and Young 1969). If ovulation rate is determined by the amount of pituitary hormone produced endogenously, then it may be possible to improve the efficiency of selection by identifying those animals that produce and utilise high levels of gonadotrophin. A search for correlates of high gonadotrophin production in the young animal, such as the size of the ovarian primordial follicle population (Mauleon and Pelletier 1964), may one day lead to selection for reproductive potential at one or two days of age. (b) The control of reproductive function The results of endocrinological research have thus far only found practical application in the use of steroid hormones to synchronise oestrus in sheep and cattle (Robinson and Lamond 1966). These techniques are characterised by variable success in the field but more consistent results should occur when the interactions of steroid and pituitary hormones are determined. As well, the development of anti-sera specific for pituitary and steroid hormones has provided a means whereby we can switch off the action of endogenous or exogenous hormones. Together with the use of synthesised releasing factors for the pituitary hormones, it would seem that we have all we need to precisely control reproductive furrctions. (c) Seasonal and lactation anoestrus Now that it is possible to monitor pituitary gland activity, a better understanding of the causes and mechanism of seasonal and lactational anoestrus should result. Information of this type will ultim ately lead to the elimination of seasonal anoestrus and a reduction in breeding interval. (d) Infertility in the young ewe Although research may show us ways to reduce the contribution of maiden ewe infertility to flock reproduction by the selection of shrewd mating strategies, an understanding of the endocrine status of young animals at various ages could lead to the development of te,chn iques to eliminate the infertil ity entirely and reduce the age at first breeding. To summarise briefly, it is contended that the rate of improvement of reproductive performance will be increased if methods designed to deal with the problems of individual producers are intensified. This would appear to be the most rapid method of realising the present reproductive potential of our domestic animals. Further, this reproductive potential may be enlarged if far sighted research is encouraged in such fields as improved and early selection for breeding potential, the precise control of reproductive functions, the elimination of seasonal anoestrus and reduction in the breeding interval, the alleviation of maiden ewe infertility 304 and earlier first breeding. All such research, however, requires an applied component necessitating co-operation between research worker and producer in order that new scientific knowledge can be effectively translated into farming practice. VI. REFERENCES A LEXANDER , G., P ETERSON, J. E., and W ATSON, R. H. (1959). Aust. vet. J. 35: 433 BEGGS, A R., and CAMPION, E. J. ( 1966). Proc. Aust. SOC. Anim. Prod. VI: 169. B IGGERS , J. D. (1968). J.Reprod. Fertil. Suppl. 8: 27. B ISHOP , A. H. ( 1964). J. Aust. Inst. agric. Sci. 30: 2 19. B ISHOP , M. W. IX .( 1964). J. Reprod. Fertil. 7: 383. B OURKE, M. E. (1967). Aust. J. exp. Agric. Anim. Hush. 7: 203. BOYD, H. ( 1965). Vet. Bu& 35: 25 1. C ONNORS , W. W. (1970). Proc. Aust. Soc. Anim. Prod. VIII: (in press). D U N, R. B. (1963). Aust. J. exp. Agric. Anim. Husb. 7: 228. D UN, R. B. (1966). Proc. N.S.W. Div. Aust. Vet. Ass. p.61. E DEY, T. (1969). Anim. Breed. Abstr. 37: 173. E DGAR, D. G. (1965). Proc. Ruakura Fmrs' Conf. Week, p.61 G ILES, J. R. (1968). Proc. Aust. Soc. A.nim. Prod. 7: 235. G ILES, J. R. (1970). M.Sc. Thesis, University of N.S.W. GODING, J., C ATT, K. J. B ROWN, J. M., K ALTENBACH, C. A., C UMMING , I. A., and M OLE B. J. (1969). Endocrinology 85: 133. H AUGHEY , K. G. (1959). Sheepfmg. A. p. 17-26. HULET , C. V., and ERCANBRACK , S. K., (1962). J. Anim. Sci. 21: 489. LIGHTFOOT, R. J., and SMITH, J. A. C. (1968). Aust. J. agric. Res. 19: 1029. L INDSAY , D. R., and ROBINSON, T. J. ( 196 1 a). J. agric. Sci., Camb. 57: 14. L INDSAY , D. R., and R OBINSON , T. J. (1961b). J. agric. Sci., Camb. 57: 137. M ANN, T. (1964). 'The Biochemistry of Semen.' (Methuen: London). M ATTNER , P. E., BRADEN, A. W. H., and T URNBULL , K. E. (1967). Aust. J. exp. Agric. Anim. Husb. 7: 103. MAULEON, P., and P ELLETIER , J. (1964). Ann. Biol. a,nim. Biophys. 4: 105. M C L AUGHLIN, J. W., RIZ~~LI, D. J., and W ILLIAMS , D. ( 1968). Proc. Aust. Soc. Anim. Prod. VII: 243. M IDGLEY , A. pi. (1967). Clin. Obst. and Gynec. 10: 119. M OORE, R. W., M C D ONALD , I. M., and M ESSENGER , J. S. ( 1966). Proc. Aust. Soc. Anim. Prod. VI: 152. M URRAY , M. (1969). Aust. vet. J. 45: 594. PEPELKO, W. E., and CLEGG, M. T. (1965a). Anim. Behav. 13: 249. PEPELKO, W. E., and C LEGG , M. T. ( 1965b). J. Anim. Sci. 24: 633. R ADFORD , H. M. (1966). Proc. Aust. SOC. Anim. Prod. VI: 19. R OBINSON, T. J., and LAMOND, D. R. (1966). Proc. Aust. Soc. Anim. Prod. VI: 10. T URNER, H. N., , and Y'OUNG, S. S. Y. (1969). 'Quantitative Genetics in Sheep Breeding.' (Macmillan : Melbourne). V AN D UIJN , C. (1965). Neth. J. agric. Sci. 13: 378. W ATSON , R. H. ( 1966). Proc. Aust. Soc. Anim. Prod. VI: 1. W ATSON, R. H., A LEXANDER, G., C UMMING, I. A. MacDoNALD, J. W., M C L AUGHLIN , J. W., RIZZOLI, D. J., and W ILLIAMS , D. ( 1968). Proc. Aust. Soc. Anim. Prod. VII: 243. W ATSON, R. H., and E LDER , E. M. ( 196 1) . Aust. vet. J. 37: 283. W ATSON, R. H., R IZZOLI, D. J., BLACKBURN, A., and R,OBINSON, D. ( 1967). J. agric. `Viet. Dep. Agric. 65: 2. WHEATLEY, 1. S., and R ADFORD, H. M. ( 1969). J. Reprod. Fertil. 19: 211. ,, 305